Patient flow from MENA to Turkey: reading search demand
A method for reading patient demand from MENA countries with Google Trends, Search Console and Keyword Planner, and which figures are not published and why.
By Roozbeh Nazari · CEO
"How many patients come to Turkey from MENA, and which treatments are they searching for?" sits on the first page of most health tourism clinics' marketing plans, and it is usually answered with second-hand figures. This article is not a list of figures. We publish no search volume, patient count or growth percentage for any country here, for a plain reason: in the environment these lines were written in, we had no access to a tool that returns search volume, and we do not write numbers we cannot verify. The aim of the article is to let you answer the same question with your own data, three free sources and a repeatable method.
Let us also say why we are this careful. Most of the demand figures circulating in health tourism come from presentations with no clear source, and clinics make language and budget decisions on the strength of them. A wrong country ranking means six months of Arabic content investment going to the wrong dialect and the wrong treatment.
Three sources, three different questions
Each of the three sources answers a different question, and they should not be used interchangeably.
- Google Trends shows direction and season, not size. The Trends FAQ says the data is a sample of real searches, normalised by dividing by the total searches of the region, and scaled to 0-100. The same page adds a warning: two regions showing the same score do not necessarily have the same search volume. So Iraq scoring higher than Saudi Arabia in Trends does not mean more searches come from Iraq; it means the topic has a larger share of all searches within Iraq.
- The Search Console performance report shows the demand your own site actually sees. The report lets you filter clicks, impressions, click-through rate and average position by country, query and page. It is the only "real" data you have, but it exists only for queries you already rank for; it will not show a treatment you have not written content about yet.
- Google Ads Keyword Planner gives monthly search estimates. Google's help page says these are estimates and that they account for seasonality. Because it works with country and language targeting, you can take a slice such as "Arabic from Saudi Arabia"; but the results are estimates and, depending on account history, may be shown as ranges.
The method: from a country list to a decision table
Step one: a query set per treatment. At most ten queries per treatment in Arabic, Persian and English; the queries are in the patient's language, not the clinic's. The Arabic set may need to be written twice to account for dialect; Gulf and Iraqi usage diverges on some treatment names. We described how this set is built in the 4-language demand map article.
Step two: in Trends, pull the last five years for each query with a country comparison. Read two things from it: season (which months peak, which months dip) and direction (rising or falling). Check in your own data whether periods such as Ramadan and the summer holidays leave a mark; write them into the content calendar after you have seen it, not before.
Step three: in Search Console, pull the last sixteen months with a country filter and put impressions, clicks and average position per country into the table. The critical reading here: countries with high impressions and weak positions are the cheapest growth area, because demand exists and Google is trying to show you, but the page is not strong enough. For countries with no impressions, Search Console is silent; fill that gap with the Planner estimate and note in the table that it is an estimate.
Step four: separate language from country. On the sites we work with, we see that a share of Arabic searches comes from European countries; those users' visa, flight and payment conditions differ from the Gulf. This slice becomes visible when you cross-filter country and page language in Search Console; if your landing page only mentions flights from the Gulf, an Arabic-speaking patient in Europe feels excluded.
Step five: do a separate reading for Persian. Demand from Iran requires a different operation because of payment and sanctions conditions, and it behaves differently on the search side too: country data in Trends is often sparser, and in Search Console Persian queries also arrive from Persian-speaking countries outside Iran. Note both gaps when writing the Persian row; "no data" and "no demand" are not the same thing.
Step six: date the table and refresh it quarterly. Because Trends data is relative, last year's score and today's score cannot be compared directly; re-pulling the same query set over the same time range at every refresh is the only method that preserves comparability. Write the pull date, the source and the filters at the top of the table; when a figure is questioned six months later, you must be able to run the same query again.
What to combine the demand table with
Search demand alone is not enough for a decision; two more pieces of data are needed on the clinic side. The first is the country-level lead and booking count in your own CRM: which country's leads turn into appointments, and which stay at the question stage. We described how to set up that measurement in the locale-based funnel in GA4 article. The second is regulation: serving international patients is subject to the Ministry of Health's authorisation regime, and the Ministry's Health Tourism Department publishes the authorised health facilities and intermediary organisations. Before opening up to a country because demand exists, whether you are on that list determines which patients you can accept.
When these two pieces of data are combined with the demand table, the decision table emerges: for each country, demand direction, your own visibility, conversion rate and operational readiness. The country ranking is the joint result of these four columns; a ranking made on Trends scores alone is, in our experience, wrong for at least one country.
Three common reading mistakes
- Treating the Trends score as volume; because of the normalisation described above, countries with small populations look inflated.
- Looking at Search Console without a country filter; impressions from Turkish queries dominate the total and the MENA signal disappears.
- Comparing countries with a single query; because of the spelling variants of Arabic treatment names, a single query measures one dialect, not the market.
Conclusion
Reading the patient flow from MENA to Turkey out of search data is possible, but only by knowing what each of the three sources does and does not say. Trends gives direction and season, Search Console gives your own real demand, the Planner gives estimates; none of them produces a country ranking on its own. We deliberately gave no figures in this article; when you fill the table with your own data, you will have a demand reading with a clear source, repeatable, and defensible in front of clinic management.